Unlocking the Language of Cells

CellWhisperer bridges the gap between transcriptomics data and natural language, enabling intuitive interaction with scRNA-seq datasets.

Decipher single cells from the
Tabula Sapiens dataset Explore the landscape of
human transcriptomes in GEO Chat with the
colonic epithelium Analyze
embryonic development
Centered Image
Spotlight paper @ ICLR 2024 MLGenX BioRxiv Code Generated datasets and models

CellWhisperer constitutes a proof-of-concept for interactive exploration of scRNA-seq data. Like other AI models, CellWhisperer does not understand user questions in a human sense, and it can make mistakes. Key results should thus be reconfirmed with conventional bioinformatics approaches.

Tutorial

(Click the annotated screenshot for a 2-minute video-tutorial)

Annotated screenshot of CellWhisperer application

Analyze your own data

To prepare your scRNA-seq data for use within CellWhisperer, follow these simple steps:

  1. Generate an h5ad file containing raw reads and a gene_name var column
  2. Run our single-command processing pipeline
  3. Start the CellWhisperer-integrated CELLxGENE Explorer
For details on these steps, refer to the "Analysis of new datasets with CellWhisperer" section in our README.

Citation

If you use CellWhisperer in your research, please cite the following preprint:

Moritz Schaefer*, Peter Peneder*, Daniel Malzl, Mihaela Peycheva, Jake Burton, Anna Hakobyan, Varun Sharma, Thomas Krausgruber, Jörg Menche, Eleni M. Tomazou, Christoph Bock (2024) Multimodal learning of transcriptomes and text enables interactive single-cell RNA-seq data exploration with natural-language chats. bioRxiv, https://www.biorxiv.org/content/10.1101/2024.10.15.618501v1

Feedback

Got feedback? Drop us an email at cellwhisperer@bocklab.org or open an issue on GitHub.